PhD Defence: Orchestrating Autonomic Software-Defined Networks for Resilience in Critical Energy Infrastructure - An application to Offshore Wind Power Plants

Just as smooth seas do not make skillful sailors, the true measure of resilience in a digital industrial network lies not in how well it performs under normal conditions but in how it responds when disruption occurs. 

Modern energy systems, such as offshore wind farms, rely on digital networks to connect, monitor, and coordinate equipment spread across large geographic areas. These networks are essential for keeping energy systems running safely and efficiently. As renewable energy infrastructure becomes more connected, its reliability increasingly depends on the health of the digital systems that support it. 
Yet greater connectivity also brings new challenges, such as technical faults, unexpected changes in network conditions, and cyberattacks that disrupt operations and affect performance. Many existing approaches rely on predefined rules, making it difficult to adapt when situations unfold in unexpected ways. 

This thesis explores a different approach to resilience, inspired by biomimicry of the human immune system. Just as the immune system continuously senses threats, coordinates responses, and adapts to changing conditions, the proposed resilience framework enables digital networks to detect problems early and respond intelligently. The framework uses cooperating adaptive agents that work together to monitor network behavior, identify disruptions, and coordinate recovery actions. 

The results show that this immune-inspired approach helps networks recover more effectively from disruptions, maintain performance under pressure, and better withstand cyber threats. By bringing adaptive, self-managing behavior into industrial networks, this research contributes to more reliable and resilient digital infrastructure for the future energy system.